Index: lib/Transforms/Utils/SimplifyCFG.cpp
===================================================================
--- lib/Transforms/Utils/SimplifyCFG.cpp
+++ lib/Transforms/Utils/SimplifyCFG.cpp
@@ -5574,10 +5574,6 @@
       V = (APInt(BitWidth, V) - Base).getLimitedValue();
   }
 
-  // Now we have signed numbers that have been shifted so that, given enough
-  // precision, there are no negative values. Since the rest of the transform
-  // is bitwise only, we switch now to an unsigned representation.
-
   // This transform can be done speculatively because it is so cheap - it results
   // in a single rotate operation being inserted.
 
@@ -5586,15 +5582,28 @@
   // 0 is the only value then a shift does nothing, and LLVM requires
   // well-formed IR to not have duplicate cases (so the minimum will not
   // be BitWidth)
+
+  // We Xor against Values[0] because the if we do not start at zero,
+  // but also don't meet the SubThreshold, then we still might share
+  // common rights bits, and if this transform succeeds...
   unsigned Shift = 64;
   for (auto &V : Values)
-    Shift = std::min(Shift, countTrailingZeros((uint64_t)V));
+    Shift = std::min(Shift, countTrailingZeros(V ^ Values[0]));
   if (Shift > 0) {
     MadeChanges = true;
     for (auto &V : Values)
       V >>= Shift;
   }
 
+  // ...then we should insert the subtraction anyways, because the rotate trick
+  // below to avoid a branch needs the shifted away bits to be zero.
+  if (Shift > countTrailingZeros(Values[0])) {
+    Base = Values[BestIndex];
+    MadeChanges = true;
+    for (auto &V : Values)
+      V = (APInt(BitWidth, V) - Base).getLimitedValue();
+  }
+
   if (!MadeChanges)
     // We didn't do anything.
     return false;
Index: test/Transforms/SimplifyCFG/rangereduce.ll
===================================================================
--- test/Transforms/SimplifyCFG/rangereduce.ll
+++ test/Transforms/SimplifyCFG/rangereduce.ll
@@ -301,12 +301,13 @@
 
 define i32 @test9(i32 %a) {
 ; CHECK-LABEL: @test9(
-; CHECK-NEXT:    [[SWITCH_RANGEREDUCE:%.*]] = call i32 @llvm.fshr.i32(i32 [[A:%.*]], i32 [[A]], i32 1)
-; CHECK-NEXT:    switch i32 [[SWITCH_RANGEREDUCE]], label [[DEF:%.*]] [
-; CHECK-NEXT:    i32 9, label [[ONE:%.*]]
-; CHECK-NEXT:    i32 10, label [[TWO:%.*]]
-; CHECK-NEXT:    i32 3, label [[THREE:%.*]]
-; CHECK-NEXT:    i32 5, label [[THREE]]
+; CHECK-NEXT:    [[SWITCH_RANGEREDUCE:%.*]] = sub i32 [[A:%.*]], 3
+; CHECK-NEXT:    [[SWITCH_RANGEREDUCE1:%.*]] = call i32 @llvm.fshr.i32(i32 [[SWITCH_RANGEREDUCE]], i32 [[SWITCH_RANGEREDUCE]], i32 1)
+; CHECK-NEXT:    switch i32 [[SWITCH_RANGEREDUCE1]], label [[DEF:%.*]] [
+; CHECK-NEXT:    i32 7, label [[ONE:%.*]]
+; CHECK-NEXT:    i32 8, label [[TWO:%.*]]
+; CHECK-NEXT:    i32 1, label [[THREE:%.*]]
+; CHECK-NEXT:    i32 3, label [[THREE]]
 ; CHECK-NEXT:    ]
 ; CHECK:       def:
 ; CHECK-NEXT:    [[MERGE:%.*]] = phi i32 [ 8867, [[TMP0:%.*]] ], [ 11984, [[ONE]] ], [ 1143, [[TWO]] ], [ 99783, [[THREE]] ]